US4303931A

US4303931A - Monolithic electroluminescent semiconductor assembly

Info

Publication number: US4303931A
Application number: US06/079,604
Authority: US
Inventors: Michel Gaffre
Original assignee: US Philips Corp
Current assignee: US Philips Corp
Priority date: 1975-09-18
Filing date: 1979-09-27
Publication date: 1981-12-01
Anticipated expiration: 1998-12-01
Also published as: GB1531897A; FR2325195B1; FR2325195A1; AU507498B2; IT1070711B; JPS5236989A; AU1773776A; DE2640518A1; CA1073548A

Abstract

A monolithic semiconductor assembly is comprised of electroluminescent diodes emitting radiations of different wavelengths. The junctions of the diodes are established in different sectors of a layer of a semiconductor compound where the concentration of a given impurity is different. The various diodes are isolated from one another.

Description

This is a division of application Ser. No. 863,456, filed Dec. 22, 1977, and which is now U.S. Pat. No. 4,198,251.

The invention relates to a monolithic electroluminescent semiconductor assembly comprising at least two electroluminescent diodes emitting two radiations of different wavelength and localised straight above two distinct areas of the surface of a semiconductor layer of a homogeneous semiconductor compound doped by at least one impurity.

Electroluminescent diodes are used for visual display notably in data processing techniques. A binary logic state may be displayed by means of an electroluminscent diode taking two states; bright diode, dark diode. However, in the case of a strong ambient light, the effects of reflection of the optics of the diodes may involve interpretation errors. For that reason it is to be preferred to visualise the various logic states by means of diodes emitting under various wavelengths, e.g. in the red and in the green. A first possibility consists in disposing side by side the necessary diodes, each of the diodes comprising its own envelope and optics. The dimensions of such a device soon become prohibitive.

In order to mitigate this said drawback, it has been tried to manufacture devices comprising two and even three diodes which are superimposed in the thickness of a composite semiconductor body formed from successive layers of different materials (or identical nominal composition but of different stoichiometry, or even differently doped), certain of the interfaces between the layers corresponding to semiconductor junctions and to the emission of radiations of specific wavelengths. Such a device which is described notably in French Pat. Xo. 2,069,256, requires a good transparency of the surface layers of the semiconductor body to the light radiation emitted by those of the junctions which are situated deep in the body. In practice it has been found on the contrary that the light absorption level is often important.

Actually, the deepest junctions are situated at several tons of microns from the emission face of the semiconductor block, because of first the thickness itself of the successive layers, and then because it is often necessary, in order to obtain a satisfactory crystal connection between the layers, to interpose between them "layers" of variable composition the presence of which increases the thickness of the structure and results in a supplementary absorption of the emitted light.

Moreover, the manufacture of devices of the type described in Patent Specification No. 2,069,256 requires, besides operations of depositing successive layers which are always delicate, localised engraving and etching operations which are also delicate in order on the one hand to give to the emitting junctions equal emissive surfaces and on the other hand to establish electric contact areas on different planes and situated outside an path of the light rays.

It is the object of the present invention to mitigate the drawbacks of the known devices such as those of the type mentioned above and to realise an assembly of electroluminescent diodes emitting different lights integrated in a same monolithic body of a plane structure, of which assembly the manufacture is easily possible by using known techniques.

The invention is based on the established fact that the emission wavelength of a luminescent semiconductor material is influenced at least partly by the impurity level which it contains.

According to the invention, an electroluminescent monolithic semiconductor assembly comprises at least two electroluminescent diodes omitting two radiations of different wavelengths, localised directly above two distinct areas of the surface of a semiconductor layer of a homogeneous semiconductor compound doped by at least one impurity and is characterized in that the junctions of the diodes are established in sectors of the layer where the concentration of the impurity is different.

Advantageously, the semiconductor layer is composed of at least two superimposed layers between which the concentrations of the impurity differ, but within which each of the concentrations remains constant at all the levels.

Since the two junctions of the semiconductor assembly are established in two sectors of the semiconductor layer where the impurity content differs, the active parts of sectors adjacent the junctions emit in bands of different wavelengths. Since the junctions are moreover situated directly above distinct areas of the surface of the semiconductor layer - advantageously chosen near each other - the drawbacks due to the superposition of emissive centers omitting radiations of different wavelengths are eliminated.

No matter what the levels are at which the junctions can be established in the thickness of the semiconductor layer, the light emitted in the direction of the outer face of the layer needs to traverse only a homogeneous material, which is an important advantage. Of course, the doping should be arranged in such manner in the layer that the light emitted from the deep levels can emerge and experience only a minimum absorption.

Since the active layer is single, the problem of the crystal connections between layers of different composition, which appears notably during the manufacture of assemblies having superimposed diodes, does not present itself with a structure according to the invention. The correlative absence of coupling layers the thicknesses of which are usually of the order of 20 to 25 μm, contribute to the decrease of the overall thickness of the structure and consequently to the decrease of light losses by absorption, which represents an important advantage of the structure.

On the other hand the assembly according to the invention presents the advantage, by its simplicity, of being easy to manufacture: it is sufficient, during the growth of the semiconductor layer, to judiciously control the supply of the doping element, as will be shown hereinafter. The constitution of the semiconductor layer itself, due to its homogeneity, is comparatively simple with respect to that of layers of complex chemical formula (for example, the layers of ternary compounds often used for the manufacture of electroluminescent devices and of which it is necessary during the growth to vary the respective proportions of two of the elements).

In addition to the specific advantages mentioned above the more general advantage should also be considered of obtaining an assembly in which the diodes, perfectly distinct because they are localised directly above distinct areas of the surface of the semiconductor layer, are well visible while excluding any risk of colour interpretation errors as it may present itself with assemblies having superimposed diodes.

The assembly according to the invention is preferably manufactured from materials which are known for their good electroluminescent qualities and are susceptible of being deposited in epitaxial layers, for example, the semiconductor compounds comprising at least one element of column III and at least one element of column V of the periodic table of elements.

In a particularly advantageous embodiment permitting obtaining diodes having different colours, going from red to green when passing through orange and yellow, the material chosen is gallium phosphide and this material is doped with nitrogen in the conditions provided by the invention.

It is known that the luminescence of gallium phosphide is susceptible to appear in various bands of visible wavelengths according to the nature of the dopant used and in particular according to the concentration of nitrogen in the material. This property has already been used advantageously for the manufacture of electroluminescent diodes or assemblies of electroluminescent diodes having monochromatic emission. However, to the knowledge of Applicant there have not yet been presented assemblies formed by nonsuperimposed integrated diodes having polychromatic emission based strictly on this same property.

Hence, as far as gallium phosphate and its dopant nitrogen are concerned, taken here as an example, it is known that an emission in the green corresponds with a concentration of the order of 10¹⁸ atoms/cm³ (the exact concentration level depends of course on other parameters, inter alia those characterising the growth conditions of the semiconductor layer while for the orange-red a concentration of approximately 10²⁰ atoms/cm³ should be reached.

On these bases Applicants propose, preferably and within the scope of the invention, on assembly manufactured on a substrate of gallium phosphide and comprising a layer of gallium phosphide obtained by epitaxy from the vapour phase (by means of this method higher concentration levels of nitrogen can be obtained than by any other method, in particular the method of epitaxy from the liquid phase), the layer being composed of at least two superimposed layers of which the first--nearest to the substrate--is doped, for example, with a nitrogen concentration of 10²⁰ atoms/cm³ throughout its thickness, and the second is doped even with 10¹⁸ atoms/cm³, also throughout its thickness, Two junctions have been formed in the structure, for example by diffusion from the surface of the second layer; one of the said junctions is situated in the thickness of the first layer and the other junction has been formed in the thickness of the second layer. So the junctions are staggered at different levels of the same semiconductor layer of a homogeneous semiconductor compound and they are established in two sectors of the layer where the nitrogen concentrations is different, as is provided by the invention.

The use of nitrogen as an impurity having a variable concentration level does not exclude the presence of a complementary doping element itself used under a fixed or variable concentration level so as to give the layer a given conductivity type. For example, with nitrogen may be combined an element selected from column VI of the periodic table, such as sulphur or tellurium, which gives the epitaxial layer an initial n-type conductivity. The junctions are then formed by diffusion of an element chosen from column II of the periodic table, for example zinc or cadmium.

The exact conditions permitting the manufacture of a device as described above, as well as the physical characteristics of the device, are given in the descriptive part of the present specification.

Of course, the said device may comprise on the same semiconductor plate an important number of diodes some omitting in a certain light spectrum, and some emitting in a spectrum different from the first. In all the cases it is advantageous to divide the semiconductor layer into insulating compartments according to a disposition well known in semiconductor technology, each compartment delimiting the domain occupied by a given diode. An absorbing layer may also be disposed between the substrate and the first epitaxial layer, which layer is destined to impede the radiation towards the rear face opposite to the semiconductor layer of the substrate, which radiation, in particular by reflection on the rear face, may traverse again the substrate, mix with the primary light and alter the purity. The simplicity of the assembly according to the invention is such that it permits the use of all known dispositions susceptable of adapting the operation to particular cases, while employing known techniques.

On the other hand it is perfectly conceivable to cause an assembly according to the invention to emit more than two distinct radiations. For three radiations, and in the embodiment described above, the semiconductor layer would thus comprise three superimposed gallium phosphide layers having a constant nitrogen concentration in each layer (and, possibly, having a constant or variable concentration of at least one other doping element), by decreasing in steps in the direction of the surface, the junctions being established in each of the layers, so for example at three different levels of the layer.

In all the cases, no matter what is the number of radiations susceptible of being generated simultaneously or alternately, it is clear that a decreasing order of the nitrogen concentration from the substrate towards the surface of the semiconductor should be respected which corresponds to an increasing order of the frequency of the emitted light vibrations, in order to avoid a sometimes total absorption of one and/or the other of radiations.

The present invention may be applied notably to the manufacture of light devices having polychromatic emission, of light signals, of display devices, in particular devices for displaying logic states.

The invention will be described in greater detail, by way of example, with reference to the accompanying drawing, in which:

FIG. 1 is a partial diagrammatic sectional view of an assembly according to the invention in an advantageous embodiment.

FIG. 2 is a partial perspective view of an integrated electroluminescent semiconductor device having bichromatic emission and comprising a number of diodes manufactured according to the invention.

The assembly shown in FIG. 1 is manufactured in a monolithic semiconductor plate formed by a substrate 1 on which an epitaxial layer 2 of a homogeneous semiconductor compound has been deposited. The layer 2 is doped by at least one impurity.

According to the invention, the layer 2 is composed of sectors, where the concentration of the the impurity is distinct. In this case, for example, two sectors are distinguished in the layer 2; a first sector corresponds to a first epitaxial layer 2A deposited on the substrate 1, on the interior of which the constant impurity concentration is situated at a first level; and a second sector corresponds to a second epitaxial layer 2B superimposed upon the first at the interior of which the equally constant impurity concentration is situated at a second level.

The epitaxial layer 2 has been separated into compartments by insulation partitions such as 3, 4, 5. According to the invention, in a first compartment limited by the

partitions

3 and 4, straight above an area of the surface of the layer 2 defined by the window 6 cut in the insulating layer 7, has been formed a first electroluminescent diode D1 of which the junction 8 is established in the first sector 2A of the layer; in a second compartment limited by

partitions

4 and 5, straight above another area of the surface of the layer 2 defined by the window 9, has been formed a second electroluminescent diode D2 of which the junction 10 is established in the second sector 2B of the layer 2.

Contact pads, produced by two

metallic deposits

11 and 12 for the diode D₁, 13 and 14 for the diode D₂, permit the provision of the necessary electrical connections with the structure.

It is noted that the two

junctions

8 and 10 are established at distinct levels of the layer 2 where, if the material of the layer is identical as regards the basic chemical formula, the concentration of doping impurities is different. The adequate use of a material and a suitable impurity results in diodes D₁ and D₂ emitting radiations of different wavelengths; for example, the deep junction 8 of the diode D₁ produces a light of dominant orange, while the surface junction 10 of the diode D₂ produces a light of medium green tone.

In the assembly as described above the diodes D₁ and D₂ are insulated in compartments. Such a disposition, already used to advantage by Applicants, notably in the structure described in the patent application of addition No. 2,080,849, is favourable, even necessary, in many cases in which the diodes must be entirely independent. In this case, however, this disposition keeps an optional character which has nothing to do with the object of the invention. According to another known disposition, there could even have been provided a light isolating layer between the substrate 1 and the sensitive layer 2 destined to avoid a light transmission in the direction of the rear face of the substrate. Moreover and according to a known embodiment of assemblies of this type, the

contacts

12 and 14 could have been replaced by a common contact layer deposited on the rear face 1A of the substrate 1, this, of course, in the case in which the material of the substrate 1 has a low resistivity. These two latter dispositions--presence of a light isolating layer and a common contact layer--also keep an optional character independent of the object of the invention.

A practical embodiment according to FIG. 1 uses gallium phosphide as a semiconductor compound and nitrogen as a main doping impurity in the operating conditions described hereinafter.

The substrate 1 is of gallium phosphide, for example having a semi-insulating character, having a resistivity between 10⁴ and 10⁷ Ωcm or even of the p-conductivity type.

On the substrate 1 which is suitably prepared the layer 2 is grown by epitaxy from the vapour phase. The operation is carried out in an apparatus of the conventional model--well known to those skilled in the art and of which it is consequently useless to give a description--in which the epitaxy furnace comprises three hot zones. In the first zone where the temperature is between 780° and 820° C. is placed gallium; there is supplied hydrogen (supply: 1500 cm³ /min.±1%) which serves essentially as a dilution gas, and hydrochloric acid (supply: 42 cm³ /min.±1%). In the second zone, or mixing zone of the gases, situated downstream of the first zone with respect to the direction of flow of the preceding hydrogen flow, heated at 950° C. (930° to 970° C.), are sent hydrogen (supply: 12000: cm³ /min.±1%) which serves as a vector gas, phosphine, PH₃, diluted to 5% in hydrogen (supply: 515 cm³ /min.±1%), diethyltelluride, Te(C₂ H₅)₂, diluted to 10 ppm in hydrogen (supply: 150 cm³, min.±1%) and ammonia. In the third zone or deposition zone which succeeds the second zone and within which is placed the substrate of gallium phosphide, the temperature is controlled between 785° and 795° C.

All figures given above remain fixed during the deposition of the whole epitaxial layer 2. Only the supply of ammonia gas varies which, for the growth of the epitaxial layer 2A, is, for example, fixed at 455 cm³ /min.±1%), which corresponds to a partial pressure of 3.2×10^-2 Atm. When the epitaxial 1 per 2A has reached the desired thickness, the ammonia supply--and only the ammonia supply--is rapidly modified to reduce it, for example, to 25 cm³ /min.±1%), for example a partial pressure of said gas being at 1.7×10^-3 Atm.; the epitaxial layer 2B is then formed.

In these conditions an epitaxial layer 2 is obtained which grows at the rate of approximately 20 μm/h and which is of the n-conductivity type. The doping level connected with the impurity telluride is 4×10¹⁶ to 2×10¹⁷ atoms/cm³ ; it is constant in the whole layer 2.

The presence of nitrogen resulting from the decomposition of ammonia gas gives a doping level of impurity of 5×10¹⁹ to 10²⁰ atoms/cm³ to the epitaxial layer 2A and of 1 to 3×10¹⁸ atoms/cm³ to the epitaxial layer 2B.

The thickness of the layer 2A is chosen between 15 and 25 μm (advantageously 20 μm), while the layer 2B has a thickness of 8 to 15 μm (preferably 10 μm). The overall thickness of the active layer 2 thus is between 23 and 40 μm.

When the layer 2 has been formed,

insulation partitions

3, 4 and 5 are formed, for example by diffusion, from the surface of the layer 2B, of an impurity of the p-type such as zinc, for example by proton bombardment.

The diodes D₁ and D₂ are then formed. For example, by a first zinc diffusion localised in the window 6 continued for 2 hours at 850° C. (845° to 855° C.) in a vacuum the deep junction 8 of the diode D₁ is formed. In a subsequent sequence of zinc diffusion carried out in the window 9 this time for 30 min. at 850° C. in a vacuum, the surface junction 10 of the diode D₂ is formed.

The depths of the diffusion forming the

junctions

8 and 10, taken between each junction and the upper limit of the epitaxial layer in which the junction is situated, are chosen to be at least equal to 1 to 3 times the diffusion length of the minority carriers in the epitaxial region in question. Given the operating conditions described before, the junction 8 is situated at 5 μm below the fictive boundary between the

layers

2A and 2B, while the junction 10 is 7 μm deep with respect to the surface of the layer 2B, this for a layer 2B having a thickness of 10 μm.

To complete the assembly, there remains the metallic deposits of the contact pads to be formed, preferably by using an alloy of gold-beryllium on the p-type regions (contacts 11 and 13) and an alloy of gold-silicon on the n-type regions (contacts 12 and 14).

The assembly as manufactured according to the operating conditions described above can emit a light having a medium yellow colour (centred on a wavelength of 0.572 μm) by its diode D₁. The diode D₂ of which the junction 10 is established in a sector of the layer 2 where the doping level of nitrogen is lower than that of the sector where the junction 8 of the diode D₁, is situated, omits green (around the wavelength of 0.560 μm). Of course, by varying the partial pressure of ammonia which determines the doping level of the layer 2 of nitrogen, the spectral characteristics of the light emitted by the manufactured diodes can be modified. Applicants have found that within the scope of the operating conditions described, only the supply of ammonia being variable, with partial ammonia pressures of 1.7×10^-3, 3.2×10^-2 and 1.2×10^-1 Atm. correspond emissions centered at 0.560 μm (green), 0.572 μm (yellow), and 0.596 μm (orange-red), respectively. Great possibilities of adjusting the light emission are connected with the adoption of intermediate partial ammonia pressures between those indicated above: with partial pressures of 6.5×10^-3 and 6.6×10^-2 correspond, for example, emissions of which the average wavelengths are situated at 0.565 μm and 0.578 μm, respectively.

It will be observed that, without great complication of the operating process described above, the supply of carrier gas of the complementary impurity (in this case tellurium) will be varied in order to modify the concentration level of the epitaxial layer of impurity and thus adjust the characteristics of the emission spectra of the diodes. If necessary, it may be arranged that the various layers of the epitaxial layer have distinct concentration levels of the complementary impurity.

In the same manner, and always in a very simple manner, it may be arranged that the concentration levels of each of the impurities--nitrogen and tellurium (or rather any impurity other than tellurium)--are variable within the same epitaxial layer.

It is interesting to note, as already indicated in the first part of the present description and as it appears clearly from FIG. 1, that the assembly according to the invention has a very simple structure, the simplicity also characterising the method of manufacturing the assembly. It should be noted in particular that the heart of the assembly is limited to a single epitaxial layer 2 of a small thickness (23 to 40 μm for a bichromatic emission) of a homogeneous material. To that effect should be disregarded the broken line 15 drawn between the

layers

2A and 2B only to materialise a transition which does not anywhere touch the material itself of the layer, but solely relates to its doping characteristics.

It is not excluded to use an assembly according to the invention by considering the light emission which is effected via the rear face 1A of the substrate 1 in so far as the substrate has a suitable transparency. In this case, however, it is clear that, if the

epitaxial layers

2A and 2B remain doped as has been indicated in the embodiment (layer 2A more highly doped with nitrogen than layer 2B), the light emitted towards the substrate 1 by the junction 10 will be absorbed, at least partly, during traversing the layer 2A and the emission will be monochromatic. So in the case of such a use, it should be ensured for the emission to remain polychromatic that the doping levels of nitrogen in the successive epitaxial layers are increasing from the substrate 1 up to the surface of the layer 2.

The device shown partly in FIG. 2 is a mosaic of electroluminescent diodes disposed in columns and arranged according to two orthogonal directions and which is made of a series of assemblies such as that of FIG. 1.

The diodes, for example 24, result from the simultaneous formation of pn-junctions in the lower part 22A of the epitaxial layer 22, while the diodes, for example 25, have their junctions--also formed simultaneously--in the surface part 22B of the layer 22. As above, the difference between the two

parts

22A and 22B resides in the doping characteristics and not in the nature and the structure of the parts. Each of the

diodes

24 or 25 is insulated in a compartment by partitions connected as a grid 23.

The connections which are not shown in FIG. 2 are realised by multilayer techniques commonly used in the manufacture of electronic microcircuits and consisting of depositing in several successive layers separated by insulating layers in the necessary regions a conductor metal according to the structure of the desired circuit in such manner as to be able to apply excitation voltages to one or several determined diodes.

It will be obvious already from FIG. 2 that the manufacturing process of such a device does not differ at all from that described for the assembly shown in FIG. 1.

Thus the invention may be supplied to any assembly comprising a plurality of electroluminescent elements formed within a monolithic body and susceptible of emitting at different light wavelengths.

Claims

What is claimed is:

1. A monolithic electroluminescent semiconductor assembly comprising

a semiconductor substrate,

a homogeneous epitaxial semiconductor layer on said semiconductor substrate and being of the same semiconductor compound as said substrate, said semiconductor layer comprising a first layer portion on said semiconductor substrate having a first impurity doping concentration and a second layer portion on said first layer portion having a second impurity doping concentration lower than that of said first impurity concentration and of the same impurity type,

at least one first electroluminescent diode junction formed in said first layer portion,

at least one second electroluminescent diode junction emitting radiation of a different wavelength formed in said second layer portion, wherein said first and second diode junctions radiate independently, and

a plurality of insulating partitions separating said first and second diode junctions.

2. A monolithic electroluminescent semiconductor assembly as in claim 1, wherein said impurity concentration is substantially constant within each of said layer portions.

3. A monolithic electroluminescent semiconductor assembly as in claim 1, wherein said impurity concentration is variable within each of said layer portions.

4. A monolithic electroluminescent semiconductor assembly as in claim 1, wherein the depth at which each junction is situated with respect to the underlying substrate of the layer portion in which it is established is equal between 1 and 3 times the diffusion length of minority carriers in the region of said junction.

5. A monolithic electroluminescent semiconductor assembly as in claim 1, wherein said semiconductor layer is doped complementarily by a second impurity whose concentration is substantially constant in said layer.

6. A monolithic electroluminescent semiconductor assembly as in claim 1, wherein said semiconductor layer is complementarily doped by a second impurity whose concentration is variable within said layer.

7. A monolithic electroluminescent semiconductor assembly as in claim 1, wherein said semiconductor substrate and homogeneous layer are gallium phosphide and said impurity is nitrogen.